|

 

Ajph129353 1.6

Published Ahead of Print on August 13, 2008, as 10.2105/AJPH.2007.129353
Cost Savings From the Provision of Specific Methodsof Contraception in a Publicly Funded Program Diana Greene Foster, PhD, Daria P. Rostovtseva, MS, Claire D. Brindis, DrPH, M. Antonia Biggs, PhD, Denis Hulett, BA, and Philip D. Darney, MD Unintended pregnancies occur increasingly Objectives. We examined the cost-effectiveness of contraceptive methods and disproportionately to women with limited dispensed in 2003 to 955 000 women in Family PACT (Planning, Access, Care and resources.1 Cost–benefit analyses have repeat- Treatment), California's publicly funded family planning program.
edly shown substantial savings to the public in Methods. We estimated the number of pregnancies averted by each contra- pregnancy-related medical expenses from the ceptive method and compared the cost of providing each method with the sav- provision of contraceptive services to low-income ings from averted pregnancies.
women.2–4 However, these analyses have not Results. More than half of the 178 000 averted pregnancies were attributable to been conducted for specific types of contracep- oral contraceptives, one fifth to injectable methods, and one tenth each to the tive methods, with the exception of a 1995 patch and barrier methods. The implant and intrauterine contraceptives were the study by Trussell et al. comparing the costs of most cost-effective, with cost savings of more than $7.00 for every $1.00 spent in using 15 different methods of contraception, services and supplies. Per $1.00 spent, injectable contraceptives yielded savings including the costs of providing the method and of $5.60; oral contraceptives, $4.07; the patch, $2.99; the vaginal ring, $2.55;barrier methods, $1.34; and emergency contraceptives, $1.43.
the costs of unintended pregnancies.5 Trussell Conclusions. All contraceptive methods were cost-effective—they saved more et al. showed the theoretical cost-effectiveness in public expenditures for unintended pregnancies than they cost to provide.
of 5 years' use of contraceptive methods, not Because no single method is clinically recommended to every woman, it is taking into account the costs of providing other medically and fiscally advisable for public health programs to offer all contra- method-related services or the likelihood of ceptive methods. (Am J Public Health. 2008;98:XXX–XXX. doi:10.2105/AJPH.
method discontinuation. Although Trussell et al.
show that all methods can be cost-effective, itis not known what the relative cost-effectivenessof specific methods is when cost data are derived Administration at no cost to the client. Methods cost of the contraceptive patch and ring.8 from an actual public health program and the available since the program's inception include However, given the absence of data on use of tendency of a significant proportion of women oral contraceptives, injectable contraceptives, these methods and the cost of providing them, to switch and discontinue methods is taken into intrauterine contraceptives, sterilization, and there has been no way to assess the validity of barrier methods. Dedicated emergency con- these criticisms. We assessed the cost-effectiveness California's family planning program, Family traceptive pills became available in 1999, and of covering new contraceptive methods for PACT (Planning, Access, Care and Treatment), the new contraceptive patch and vaginal ring women aged 13 to 44 years and evaluated provides contraception and reproductive were added to the formulary in 2002. With the relative contribution of all methods to the health services to women and men of repro- the introduction of new contraceptive methods, fertility effect of the Family PACT Program.
ductive age whose incomes do not exceed the pattern of methods dispensed through We compared the costs of providing contra- 200% of the federal poverty level and who Family PACT has changed: by 2005, nearly ceptives through Family PACT with the costs have no other reproductive health care cover- 20% of women served each year received at of unintended pregnancies to government age. More than 2000 private and nonprofit least 1 pack of emergency contraceptives, providers across the state deliver family plan- 15% received a contraceptive patch, and 2% ning services and are reimbursed by the received a contraceptive ring, whereas the Family PACT program on a fee-for-service percentage of women receiving oral and in- basis.6 The program was launched in 1997 and jectable contraceptives has slightly declined.
We employed the same methodology to grew rapidly, serving 750000 clients during The percentage of women receiving barrier estimate pregnancies averted that we used its first full year of operation and more than methods with or without another contraceptive to estimate the fertility effect of the Family 1.6 million per year in recent years.7 The size has remained steady at around 45%.7 PACT Program in fiscal years 1997–1998 of the program and the detailed data kept by the Given their limited time on the market and and 2002.9,10 For this study, we analyzed the program on contraceptive methods dispensed a lack of large-scale data on contraceptive fertility effect of each method of contraception permit an analysis of the cost-effectiveness of method dispensing, little is known about the separately. We estimated the number of preg- specific methods of contraception.
women using the new methods and the effect nancies averted through the use of specific con- Family PACT covers all contraceptive of these methods on unintended pregnancy.
traceptive methods to be the difference between methods approved by the Food and Drug There has been some criticism of the high the number of pregnancies expected in the October 2008, Vol 98, No. 10 American Journal of Public Health Foster et al. Peer Reviewed Research and Practice 1 RESEARCH AND PRACTICE absence of that method and the number ex- Estimating the Number of Pregnancies were assumed to have the same probability of pected given the provision of that method failure as oral contraceptives. We assumed that through Family PACT.
We estimated the probability of pregnancy 40% of pregnancies end in induced abortion, in the absence of each method of contraception 10% end in spontaneous abortion, 1% are ec- Contraceptive Coverage as follows. For each woman who received a topic, and the remaining 49% are carried to For this study we relied on 3 full calendar new contraceptive method in 2003, we looked years of Family PACT claims data. We used at the previously used method and calculated We expected that in the absence of each contraceptive method dispensing claims data her probability of pregnancy in the absence method of contraception, some women would from 2003. We also examined claims in 2002 of the new method. For example, to estimate use less effective methods and some would use to identify methods previously dispensed to the fertility rate in the absence of the contra- more effective methods. We assumed that a clients who received methods in 2003 and ceptive patch, we looked at the last methods woman who adopted a method of contracep- claims in 2004 to predict contraceptive con- dispensed to patch users prior to their 2003 tion at her first Family PACT visit would in the tinuation, including intrauterine device and patch visit. For women who had no previous absence of the Family PACT program, return to implant removals.
Family PACT visits or who had not received using the method she used prior to enrollment.
We estimated the number of months of contraceptives from Family PACT in the pre- In the absence of individual-level data contraceptive coverage provided under Family vious year, we used a programwide fertility rate on methods used prior to enrollment, we used PACT on the basis of paid claims data on the estimated from clients' self-reports of contra- data from women new to Family PACT for quantity and type of contraceptives dispensed.
ceptive use prior to enrollment. These self- whom contraceptive method use prior to en- The coverage for long-term methods (tubal reports were taken from a review of medical rollment was abstracted as part of a medical ligations, intrauterine devices, and implants) records for dates of service in fiscal year record review. Therefore, we anticipated that was calculated as the number of months 2000–2001 of 868 new Family PACT female a method dispensed during a client's first visit between the provision date and December clients who were not pregnant and not seeking would have an associated pregnancy rate in 2004, unless the claims data showed an im- the absence of that method equal to that of plant or intrauterine device removal before To estimate the number of pregnancies the absence of the program.
December 2004. We imposed this 2-year among clients, we modeled the month-by- cap to avoid predicting pregnancies far into month experience of each woman who re- Costs of Providing Services ceived a contraceptive method, beginning To calculate the costs of providing contra- Because clients may not use all of the con- with the month when the contraceptive was ceptive services by method, we assigned a traceptives they receive, we adjusted the num- dispensed and ending with the last month of primary method to each client, on the basis of ber of months of contraceptive coverage for contraceptive coverage. For each month, we what was dispensed at each visit, and all sub- short-term methods, such as condoms and oral used a Markov model to calculate the proba- sequent expenses for that client were attributed contraceptives, to account for method discon- bility that the woman would become pregnant to that method until she was given a different tinuation. For oral contraceptives, we assumed on the basis of the failure rate of the method method. In assigning the primary method, for that a woman who did not return for refills used (i.e., the proportion of users who experi- clients who were dispensed more than 1 used half of the pills dispensed to her. We as- ence a pregnancy in a year), age-specific fe- method, we used the following rank-ordered sumed that women who received 1 packet of cundity, and the estimated probability of preg- list: tubal ligation, intrauterine contraceptive, emergency contraceptive pills used it; however, nancy in the months before the contraceptive implant, injectable, vaginal ring, patch, oral for women who were given more than 1 packet was dispensed. Modeling pregnancies averted contraceptives, barrier method, emergency (provision in advance), we assumed that 50% by month allowed us to use specific contra- contraceptive pill. A client who received an used the second packet. We assumed a month ceptive dispensing data for months of coverage, implant and a ring was assigned the implant as of protection for every 12 condoms dispensed rather than assuming a year of coverage for her primary method, because implant comes by pharmacies. For condoms and other barrier each client. It also allowed for repeat pregnan- before ring in the list. In addition to the cost methods dispensed by clinics, the exact quan- cies within a year, a common outcome among of the contraceptive supplies themselves, the tity of supplies dispensed was not available, and women who use low-efficacy methods and costs of all subsequent clinician visits, labora- we assumed, given findings from the Family terminate pregnancies by abortion.9 tory work, and pharmacy claims, including PACT medical records review, that each dis- For this analysis, we made the same as- services related to pregnancy testing or sexu- pensing provided 2 months of contraceptive sumptions about contraceptive failure rates ally transmitted infections, for a client were coverage. Each injection was assumed to pro- and pregnancy outcomes as in our study of the attributed to her primary method. For clini- vide 3 months of contraceptive coverage. In fertility effect of the entire Family PACT Pro- cian visits, laboratory work, or pharmacy our sensitivity analysis, we examined the effect gram.9,10 The monthly probability of pregnancy visits that did not involve the dispensing of of adjusting for method discontinuation on our by method was derived from typical-use annual contraceptives, costs were assigned to the estimates of method-specific pregnancies probability of pregnancy by method from primary method of the client's last clinician Hatcher et al.11 The contraceptive patch and ring 2 Research and Practice Peer Reviewed Foster et al.
American Journal of Public Health October 2008, Vol 98, No. 10 RESEARCH AND PRACTICE In 2003, Family PACT spent $385 million the absence of Family PACT, would use no contraception and 3% were using natural on female clients. Of that amount, $47 million method. In our second sensitivity analysis, all family planning methods such as periodic (12%) was spent on clients who were not given methods had the same probability of preg- abstinence and withdrawal. Nearly 4 in 10 a contraceptive method by the program in nancy in the absence of their use. Third, we women (38%) were using condoms, and the 2003, $28 million was spent on clients before examined the short-term financial returns remaining 32% were using hormonal methods their first dispensing visit of 2003, and $8 of contraceptive provision, examining the or intrauterine devices. Girls and women aged million was spent on clients outside the age medical savings only from the time of concep- 13 to 19 years were more likely than were range of our study. The remaining $302 mil- tion to the time of delivery or termination.
women aged 20 to 44 years to be using lion we attributed to particular primary condoms and less likely to be using hormonal methods of contraception for the purposes or intrauterine contraceptives. We estimated of our study.
that 43% of the women would have become Contraceptives Dispensed to Women in pregnant over the course of a year in the Costs of Unintended Pregnancies absence of Family PACT. Our estimates of As part of a separate cost–benefit study,2 Nearly 1 million female clients—217 000 pregnancy rates in the absence of each we estimated the cost to the public of an unin- aged 13 to 19 years and 738 000 aged 20 method of contraception are shown tended pregnancy for 2 years after the birth.
to 44 years—received contraceptive methods In 2002, each pregnancy averted by contracep- through Family PACT in 2003. Payments tion provided by Family PACT that would were made for oral contraceptives for about Pregnancies Averted by Specific have ended in abortion saved the public sector 449 000 clients, condoms and other barrier Methods of Contraception in 2003 $372. Each birth averted would have cost the methods as a primary method for 405 000 On the basis of the quantity and type of public $3228 from the time of conception to clients, injectables for 162 000 clients, and contraceptive methods dispensed, we esti- the time of delivery and $11545 in medical, long-term methods for 22 000 clients. Emer- mated that because of method failure and welfare, and other social service costs for a gency contraceptives were dispensed without noncompliance, Family PACT clients experi- woman and child from the time of conception any other method to 37 000 clients. About enced 37 000 pregnancies during the time until 2 years after the birth.2 We adjusted the 129 000 women received the new contracep- they were using contraceptives dispensed in costs for pregnancies leading to birth to factor tive patch, and 11000 received the vaginal 2003. If these women had been using the in the estimated 38% of adolescent pregnan- contraceptive ring.
methods they used before adopting a primary cies and 50% of adult pregnancies that were The claims paid for women during 2003 method through Family PACT, or for women merely delayed, rather than prevented, by con- provided each client with an average of 6.6 with no history of contraceptive use, the same traceptive use and that would still result in costs to months of primary-method contraceptive cov- method array as women new to the program, the public when they occurred at a later date.2 erage. Oral contraceptives accounted for half they would have experienced 216 000 preg- (50%) of the woman-months of protection nancies. The difference, 178 000 pregnan- Sensitivity Analyses dispensed, followed by injectables (17%), bar- cies, is an estimate of the number of preg- Our model of the cost savings from pre- rier methods (14%), and the patch (11%).
nancies averted through the provision of venting unintended pregnancies with specific Users of long-term methods (tubal ligation, specific contraceptive methods by Family methods of contraception made some assump- intrauterine devices, and implants) received PACT in 2003 (Table 2). This estimate is tions that might have affected the relative cost- the greatest number of months of protection.
somewhat lower than the estimate of effectiveness we found for particular methods.
Among reversible, short-term methods of con- 205 000 pregnancies averted by the program We conducted 3 sensitivity analyses to inves- traception, oral contraceptives provided the in 200210 because in this analysis some women tigate the results' dependence on the probabil- greatest number of months of protection (7.1 continue to use subsidized contraceptive ser- ity estimates chosen. First, in our base model months), followed by injectables (6.4 months) vices, they just would revert to methods they had we adjusted the months of protection from short- and the patch (5.5 months). Barrier methods previously used in the program.
term methods of contraception to account for provided 2.2 months of protection and emer- When we estimated pregnancies averted method discontinuation. As a sensitivity anal- gency contraceptives used as the primary by method, we found that slightly more than ysis, we present our findings without this ad- method provided 1.3 months, although clients half (91000) of the averted pregnancies were justment, assuming that clients used all the who used these methods may have received attributable to oral contraceptive use, 22% supplies they were given. Second, our use of other primary methods over the course of the (39 000) were attributable to injectable con- estimated method-specific pregnancy rates in year (Table 1).
traceptive use, 10% (18 000) were attrib- the absence of each method of contraception, utable to the contraceptive patch, approxi- which took into account the previous methods Pregnancy Rates in the Absence of mately 10% (17 000) were attributable to use used, made some methods appear to be rela- Family PACT or Specific Methods of barrier methods, and 6% (11000) were tively more effective at reducing pregnancies, Before enrolling in Family PACT, 27% of attributable to use of long-term methods particularly those adopted by women who, in the women were using no method of October 2008, Vol 98, No. 10 American Journal of Public Health Foster et al. Peer Reviewed Research and Practice 3 RESEARCH AND PRACTICE TABLE 1—Primary Contraceptive Methods Provided to Female Family PACT Clients and Estimated Pregnancy Rates in the Absence of the Method: California, 2003 Clients Aged 13–19 y (n = 217 263) Clients Aged 20–44 y (n = 738 331) All Clients (n = 955 594) Estimated Pregnancy Average Months Estimated Pregnancy Percentage of Total Months of of Protection per Primary Contraceptive Protection Attributable to Method Interval tubal ligationb Intrauterine deviceb Oral contraceptives Emergency contraceptives Note. PACT = Planning, Access, Care, and Treatment.
aNumbers in this column add to more than the n for this group because some women made visits for more than 1 primary method over the course of the year.
bAssumes a 2-year cap on duration of contraceptive protection.
Cost Savings per Dollar Spent on Family produced the second highest savings per dollar contraceptives ($1.43) produced lower savings PACT by Primary Contraceptive Method spent, despite the fact that we capped the per $1.00 spent on services.
All contraceptive methods were cost-effective duration of contraceptive protection it pro- in that they prevented unintended pregnancies, vided at 2 years. Among short-term methods, Sensitivity of Results to Methodology but the cost savings per dollar spent varied by injectable contraceptives produced the highest method (Table 2). The contraceptive implant savings, at $5.60 saved per $1.00 spent on In our first sensitivity analysis, we examined produced the highest savings; however, the services, followed by oral contraceptives the number of pregnancies averted estimated small number of implant users may have ($4.07), the patch ($2.99), and the ring ($2.55).
without adjustment for discontinuation of bar- skewed these data. The intrauterine device Barrier methods ($1.34) and emergency rier methods, the ring, the patch, and oralcontraceptives. This adjustment reduced themonths of protection provided by these TABLE 2—Number of Pregnancies Averted and Cost Savings for Each Contraceptive Method methods and reduced the cost savings associ- Provided by Family PACT: California, 2003 ated with their use. We examined whetheradjustment for discontinuation of use dispro- Costs Associated With portionately affected one short-term method Pregnancies With Pregnancies Provision of Method, more than others. As Table 3 shows, the cost Absence of Method Dollar Expenditure, $ savings associated with use of barrier Interval tubal ligationa methods is disproportionately reduced by this adjustment. However, even assuming Intrauterine devicea that clients used all the supplies they re- ceived does not increase the savings for barrier methods to the level of savings for hormonal methods.
Oral contraceptives In our second sensitivity analysis, we exam- ined the effect of our method-specific failure Emergency contraceptives rates for the absence of each contraceptive method. We compared the results we obtainedby using method-specific failure rates with Note. PACT = Planning, Access, Care, and Treatment.
aAssumes a 2-year cap on duration of contraceptive protection.
1 programwide rate of unintended pregnancyin the absence of the program. The 4 Research and Practice Peer Reviewed Foster et al.
American Journal of Public Health October 2008, Vol 98, No. 10 RESEARCH AND PRACTICE TABLE 3—Sensitivity Analyses of Method-Specific Savings From Preventing Unintended Pregnancies Scenario 1: Women Use All Scenario 2: Constant Risk of Pregnancy Across Scenario 3: Medical Costs Through Methods Dispensed Methods in Absence of Program End of Pregnancy Only Savings per Dollar Cost Savings per Dollar Percentage Change Percentage Change Percentage Change Dollar Expenditure, $ Dollar Expenditure, $ Interval tubal ligation Intrauterine device Oral contraceptives Emergency contraceptives programwide rate of pregnancy was higher on contraceptives provided through the pro- because fewer women used contraceptive gram was associated with only $1.06 in savings.
methods in the absence of the family planning Only the ring, the patch, emergency contra- program. If all women adopted the methods ceptives, and barrier methods were not asso- We made assumptions that had the effect of that new Family PACT clients used prior to ciated with positive savings within 9 months reducing the cost savings associated with very- their enrollment, the savings for all contracep- of conception.
long-term and very-short-term contraceptive tive methods under Family PACT would be The cost-efficiency curve in Figure 1 shows methods. The conservative cap of 2 years on almost 19% higher.
the investment in contraceptive services per the effect of long-term methods resulted in an In our third sensitivity analysis, by limiting user along the x-axis and the effectiveness in underestimation of the cost savings from these the cost associated with unintended pregnan- pregnancies averted along the y-axis. At low methods, which women may continue to use cies to just those medical expenditures that levels of investment there were slim returns in for many years. We may also have under- occur up to the time of delivery or termination, effectiveness. Very high levels of investment estimated the duration of contraceptive cover- we obtained a conservative measure of the per user yielded diminishing returns. Invest- age for barrier methods if clients' supplies short-term returns of providing contraception.
ments in intrauterine devices and implants lasted longer than 2 months or overestimated Within 9 months of conception, $1.00 spent yielded the highest returns.
the duration of coverage if they ran out before2 months.
We did not capture the cost savings associ- ated with postpartum tubal ligations, becauseFamily PACT covers only interval tubal liga-tions (procedures not done in conjunction withhospitalization for delivery). Postpartum tuballigations are likely associated with higher costsavings than interval tubal ligations.
Our implant provision costs and use are for Norplant devices that were provided in 2003,when the product was no longer on the marketbut doctors were still implanting supplies theyhad in stock. However, the latest implant de-vice, Implanon, would likely have similar effi-cacy and continuation in the first 2 years ofuse. The cost of the Implanon device is about FIGURE 1—Cost-efficiency of contraceptive methods provided by California's Family PACT 20% higher than the cost of Norplant, which may (Planning, Access, Care, Treatment) program, 2003.
translate into slightly lower cost-effectivenessthan we estimated for Norplant.
October 2008, Vol 98, No. 10 American Journal of Public Health Foster et al. Peer Reviewed Research and Practice 5 RESEARCH AND PRACTICE Our study year, 2003, was the first full year decisions. For new users of any contraceptive Family PACT program report fiscal year 06/07.
in which the contraceptive patch and ring were method, follow-up support should be available A report to the State of California Department of PublicHealth Office of Family Planning. Available at: http:// available through Family PACT. As a result, to ensure the user's compliance and under- many users were likely first-time users who standing and to ascertain the method's accept- 2006-07.pdf. Accessed July 28, 2008.
may have been given a small supply on a ability. Users of barrier methods and emer- Family PACT program report fiscal year 04/05.
trial basis. With the passage of time, clients gency contraceptives should be encouraged A report to the State of California Department ofHealth Services Office of Family Planning. Available may be given a larger quantity of these prod- to use additional, longer-term methods of con- ucts, perhaps providing coverage equivalent to traception. Together, these measures will con- GraphicSummaryFY04-05.pdf. Accessed June 26, oral contraceptives, which would increase the tribute to higher contraceptive compliance cost savings associated with providing and continuation, lower failure rates, and Sonfield A. Summer price spike: a case study about publicly funded clinics and the cost of contraceptive these new methods of contraception.
fewer unintended pregnancies. j supplies. Guttmacher Policy Rev. 2006;9:2–5.
Foster DG, Klaisle CM, Blum M, Bradsberry ME, Brindis CD, Stewart FH. Expanded state-funded family We found all contraceptive methods dis- planning services: estimating pregnancies averted bythe Family PACT Program in California, 1997–1998.
pensed through Family PACT to be cost-effective.
About the Authors Am J Public Health. 2004;94:1341–1346.
Long-term methods are very cost-effective.
All authors are with the Bixby Center for Global Repro- 10. Foster DG, Biggs MA, Amaral G, et al. Estimates of ductive Health, University of California, San Francisco.
Barrier methods and emergency contraceptives pregnancies averted through California's family planning Requests for reprints should be sent to Diana Greene tend to yield the lowest savings per dollar spent waiver program in 2002. Perspect Sex Reprod Health.
Foster, PhD, 1330 Broadway, Suite 1100, Oakland, CA because of their relatively low efficacy and 94612 (e-mail: [email protected]).
This article was accepted May 26, 2008.
11. Hatcher RA, Trussell J, Nelson AL, et al. Contra- short duration of use. Higher costs and fewer ceptive Technology. 18th ed. New York, NY: Ardent months of contraceptive protection from the Media; 2004.
contraceptive patch and ring result in lower 12. Henshaw SK. Unintended pregnancy in the United D. G. Foster originated the study, carried out the anal- cost savings than for oral contraceptives; States. Fam Planning Perspect. 1998;30:24–29.
ysis, and drafted the article. D. P. Rostovtseva and M. A.
however, these easier-to-use methods offer Biggs assisted with the development of the study and the 13. Saraiya M, Berg CJ, Shulman H, Green CA, Atrash writing of the article. C. D. Brindis and D. P. Darney HK. Estimates of the annual number of clinically recog- enhanced convenience for clients. Highly user- oversaw the project and provided key advice and editing.
nized pregnancies in the United States, 1981–1991.
dependent methods, such as oral contracep- D. Hulett oversaw the data abstraction and interpreta- Am J Epidemiol. 1999;149:1025–1029.
tives and barrier methods, are less effective tion. All authors reviewed the article.
14. Rosenberg MJ, Waugh MS, Burnhill MS. Compli- because they are more likely to be used im- ance, counseling and satisfaction with oral contraceptives:a prospective evaluation. Fam Plann Perspect.
perfectly. Missed oral contraceptive pills are This research was funded by the State of California Office quite common and contribute to unintended of Family Planning through a contract to the Bixby 15. Ornstein RM, Fisher MM. Hormonal contraception pregnancy.14–16 Recent research indicates that Center for Global Reproductive Health, University of in adolescents: special considerations. Paediatric Drugs.
compliance rates are at least as high for the ring California, San Francisco.
as for oral contraceptives and that patch com- 16. Rosenberg MJ, Waugh MS, Long S. Unintendedpregnancies and use, misuse and discontinuation of pliance may be even higher, which may lead to Human Participant Protection oral contraceptives. J Reprod Med. 1995;40:355–360.
This research was approved by the University of Cal- lower failure rates and greater cost savings.17,18 ifornia, San Francisco Committee for Human Research 17. Archer DF, Cullins V, Creasy GW, Fisher AC. The This would be especially true if women who (CHR H429-16233).
impact of improved compliance with a weekly contra- use the patch or the ring receive sufficient ceptive transdermal system (Ortho Evra) on contracep-tive efficacy. Contraception. 2004;64:189–195.
quantities to provide coverage equal to that 18. Oddsson K, Leifels-Fischer B, de Melo NR, et al.
provided by oral contraceptives.
Finer LB, Henshaw SK. Disparities in rates of Efficacy and safety of a contraceptive vaginal ring Because all contraceptive methods are cost- unintended pregnancy in the United States, 1994 (NuvaRing) compared with a combined oral contracep- effective, public health programs can offer a and 2001. Perspect Sex Reprod Health. 2006;38:90–96.
tive: a 1-year randomized trial. Contraception.
2004;71:176–182.
range of methods to increase the chances that Amaral G, Foster DG, Biggs A, Jasik C, Judd S, Brindis C. Public savings from the prevention of unin- their clients will find a method that suits their tended pregnancy: a cost analysis of family planning needs. Providers should be encouraged to services in California. Health Serv Res. 2007;42: dispense or prescribe more months of con- traceptive protection per visit as appropriate, Forrest JD, Samara R. Impact of publicly funded contraceptive services on unintended pregnancies and which would reduce the number of clinic implications for Medicaid expenditures. Fam Plann Per- visits and costs while increasing method con- Forrest JD, Singh S. The impact of public-sector Women using family planning services expenditures for contraceptive services in California.
Fam Plann Perspect. 1990;22:161–168.
should be given information about the rela- Trussell J, Leveque JA, Koenig JD, et al. The tive effectiveness of different contraceptive economic value of contraception: a comparison of methods so that they can make educated 15 methods. Am J Public Health. 1995;85:494–503.
6 Research and Practice Peer Reviewed Foster et al.
American Journal of Public Health October 2008, Vol 98, No. 10

Source: http://www.ansirh.com/_documents/library/foster_ajph10-2008.pdf